Maybe not a court order. But I could get behind a process similar to other utilities where you have months or warning and paperwork.
I believe the saying goes.
If you make something idiot proof, the universe will produce a more advanced idiot.
(Not to belittle your late father’s intellect)
Never done this to a single server.
Managed to write the “ufw enable, deny all” part of ansislbe script without the “allow 22” part and run it against all my homelab once.
I swear this is the nerds fault. We showed the Kubernets “Hey look, if we need more we can make more. If we need less we can just turn off the ones we don’t need automatic” and some manager is like “I can do that”
Right? And it’s so easy to say from the start “Yeah, one of the Artist used some AI on some background items.” And just own up to it. AI is at the level of ‘Everyone is doing it’. All you have to do is make some vague promises of “We’ll try and do better and catching poorly done AI in the future” and everyone would just shrug and nod.
It’s the unforced error of lying about it that’s the bad part.
Tiny devils advocate, IF we can make it so ONLY Google can spy on us and malware adware can NOT spy on us would be an “improvement”. Google is a lot easier to target with regulation and stuff.
That said, I wouldn’t touch Google with a 10 foot pole.
I didn’t know DARPA was a thing since it made the internet